mariodefelipe/sap-bdc-mcp-server

本页由 AI 翻译，技术细节请以原文为准。查看原文

11 stars 5 forks 质量 71 趋势 18MIT

MCP server for SAP Business Data Cloud (BDC) integration - 7 tools for Delta Sharing, data product publishing, validation, and orchestration

概览

An MCP (Model Context Protocol) server that provides integration with SAP Business Data Cloud (BDC) Connect SDK. This server enables AI assistants like Claude to interact with SAP BDC for data sharing, Delta Sharing protocol operations, and data product management. : ✅ Released on PyPI - (2026-01-10) : ✨ Added provision_share (end-to-end orchestration) and validate_share_readiness (pre-flight validation) tools! This MCP server provides for SAP BDC operations: - : Manage data shares with ORD metadata - : Configure shares using Common Semantic Notation - : Publish and unpublish data products - : Remove and withdraw shared resources - : Auto-generate CSN templates from Databricks shares - ✨: One-step share creation, granting, and registration - ✨: Validate shares before registration to prevent errors - Python 3.9+ (Python 3.

README

以下为原文 README 节选，完整内容请查看原文。

SAP Business Data Cloud MCP Server

Status: ✅ Released on PyPI - v0.2.0 (2026-01-10)

New in v0.2.0: ✨ Added provision_share (end-to-end orchestration) and validate_share_readiness (pre-flight validation) tools!

Features

This MCP server provides 7 powerful tools for SAP BDC operations:

Create/Update Shares: Manage data shares with ORD metadata
CSN Schema Management: Configure shares using Common Semantic Notation
Data Product Publishing: Publish and unpublish data products
Share Deletion: Remove and withdraw shared resources
CSN Template Generation: Auto-generate CSN templates from Databricks shares
End-to-End Provisioning ✨: One-step share creation, granting, and registration
Pre-flight Validation ✨: Validate shares before registration to prevent errors

Prerequisites

Python 3.9+ (Python 3.11+ recommended for local development)
Access to a Databricks environment
SAP Business Data Cloud account
Databricks recipient configured for Delta Sharing
For local development: Databricks personal access token

Quick Start

Installation

Choose your preferred language/platform:

Python (PyPI)

pip install sap-bdc-mcp-server

Node.js/TypeScript (npm)

npm install @mariodefelize/sap-bdc-mcp-server

Note: The npm package requires Python 3.9+ to be installed, as it wraps the Python MCP server.

See NPM_README.md for full Node.js/TypeScript documentation.

From Source

# Clone the repository
git clone https://github.com/MarioDeFelipe/sap-bdc-mcp-server.git
cd sap-bdc-mcp-server

# Install Python package in development mode
pip install -e .

# Install npm dependencies (optional, for Node.js development)
npm install

Configuration

For Local Development (Recommended)

Create a .env file in the project root:

# Databricks Configuration
DATABRICKS_RECIPIENT_NAME=your_recipient_name
DATABRICKS_HOST=https://your-workspace.cloud.databricks.com
DATABRICKS_TOKEN=your_databricks_token

# Optional
LOG_LEVEL=INFO

The server will automatically use LocalDatabricksClient which works without dbutils.

For Databricks Notebook Environment

If running inside Databricks notebooks, only set:

DATABRICKS_RECIPIENT_NAME=your_recipient_name
LOG_LEVEL=INFO

在 GitHub 查看完整 README

安装

This server does not publish a one-line install command.

Open the repository installation guide 新手？先看 mcpskill.org 教程 →

配置

{
  "mcpServers": {
    "sap-bdc": {
      "command": "python",
      "args": ["-m", "sap_bdc_mcp.server"],
      "env": {
        "DATABRICKS_RECIPIENT_NAME": "your_recipient_name"
      }
    }
  }
}